University of Cambridge > Talks.cam > Microsoft Research Cambridge, public talks > HyperDex: A Consistent, Fault-tolerant, Searchable NoSQL Store

HyperDex: A Consistent, Fault-tolerant, Searchable NoSQL Store

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Microsoft Research Cambridge Talks Admins.

Distributed key-value stores are now a standard component of high-performance web services and cloud computing applications. While key-value stores offer significant performance and scalability advantages, the first wave of NoSQL stores typically compromise on consistency and drastically limit the system API .

In this talk, I will present HyperDex, a novel, open-source distributed key-value store that provides (1) strong consistency guarantees, (2) fault-tolerance for failures and partitions affecting up to f nodes, and (3) a rich API which includes a unique search primitive that enables queries on secondary attributes. HyperDex achieves these properties through the combination of two recent technical advances called hyperspace hashing and value-dependent chaining. Performance measurements from the industry-standard YCSB benchmark show that HyperDex these properties do not extract a high overhead: HyperDex is actually a factor of 2-13 faster than Cassandra and MongoDB.

I will also briefly present two of our on-going projects on improving NoSQL stores. The first project is a new lightweight distributed transaction processing framework for web services. This framework avoids multi-phase commit protocols and coordinator election by organizing transaction participants into transactional chains. The second project is a new NoSQL store that offers both performance isolation between jobs and high system utilization through reactive, deadline-driven replication and an affinity-based job scheduling scheme

Overall, this talk outlines our vision on what properties the next generation data stores should offer their users, and the steps we are taking towards efficiently providing these properties in our systems.

This talk is part of the Microsoft Research Cambridge, public talks series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2024 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity